GeForce GTX 660M vs Tesla M2090

NVIDIA

GeForce GTX 660M

2012Core: 835 MHzBoost: 950 MHz
VS
NVIDIA

Tesla M2090

2011Core: 651 MHz

GeForce GTX 660M vs Tesla M2090 Performance Spectrum

About G3D Mark

G3D Mark is a standard benchmark that measures graphics performance in real-world gaming scenarios. It simplifies comparing cards from different brands, where higher scores directly correlate with better fps and smoother gaming experiences.

GeForce GTX 660M vs Tesla M2090: Pros, Cons & Final Verdict

See where each GPU makes more sense in practice: raw FPS, VRAM, features, power draw, pricing, and long-term headroom.

GeForce GTX 660M

2012

Why buy it

  • 300% more VRAM for high-resolution textures and newer games (2 GB vs 512 MB).
  • Draws 50W instead of 250W, a 200W reduction.
  • More future proof: Kepler (2012−2018) on 28nm with a newer platform for upcoming games.

Trade-offs

  • 2012 hardware with 2 GB of VRAM is already well past its comfortable zone for modern gaming, so it is hard to recommend now.
  • Lower G3D Mark per dollar, at 0 vs 0.6 G3D/$ (Unknown MSRP vs $2,500 MSRP).

Tesla M2090

2011

Why buy it

  • Delivers 100+% more G3D Mark for each dollar spent, at 0.6 vs 0 G3D/$ ($2,500 MSRP vs Unknown MSRP).

Trade-offs

  • Less VRAM, with 512 MB vs 2 GB for high-resolution textures and newer games.
  • 2011 hardware with 512 MB of VRAM is already well past its comfortable zone for modern gaming, so it is hard to recommend now.
  • 400% higher power demand at 250W vs 50W.

Quick Answers

Which GPU is faster for gaming right now?
GeForce GTX 660M is the faster gaming card right now based on the synthetic data we have. It leads by 3.7% in PassMark G3D (1,452 vs 1,400), which is the best performance signal available in this matchup.
Which GPU is the safer long-term pick for 2026 and beyond?
GeForce GTX 660M is the safer long-term pick for 2026 and beyond. The case is simple: 2 GB vs 512 MB of VRAM, a 28nm process instead of 40nm, and a newer 2012 generation instead of 2011. That gives it more room for heavier textures and higher settings over time.
Which GPU is the better buy today?
Tesla M2090 makes the most sense today based on the pricing and value data we have for this matchup. If you are mainly targeting 1080p and some 1440p, Tesla M2090 is the easier value choice. If you care more about 1080p and some 1440p headroom, GeForce GTX 660M has the stronger long-term case.

GeForce GTX 660M vs Tesla M2090 Technical Specifications

Side-by-side specs, architecture details, clocks, memory, power, and platform differences.

NVIDIA

GeForce GTX 660M

The GeForce GTX 660M is manufactured by NVIDIA. It was released in March 22 2012. It features the Kepler architecture. The core clock ranges from 835 MHz to 950 MHz. It has 384 shading units. The thermal design power (TDP) is 50W. Manufactured using 28 nm process technology. G3D Mark benchmark score: 1,452 points.

NVIDIA

Tesla M2090

The Tesla M2090 is manufactured by NVIDIA. It was released in July 25 2011. It features the Fermi 2.0 architecture. The core clock speed is 651 MHz. It has 512 shading units. The thermal design power (TDP) is 250W. Manufactured using 40 nm process technology. G3D Mark benchmark score: 1,400 points.

Graphics Performance

The GeForce GTX 660M scores 1,452 and the Tesla M2090 reaches 1,400 in the G3D Mark benchmark — just a 3.7% difference, making them near-identical in rasterization performance. The GeForce GTX 660M is built on Kepler while the Tesla M2090 uses Fermi 2.0, both on 28 nm vs 40 nm. Shader units: 384 (GeForce GTX 660M) vs 512 (Tesla M2090). Raw compute: 0.7296 TFLOPS (GeForce GTX 660M) vs 1.332 TFLOPS (Tesla M2090).

FeatureGeForce GTX 660MTesla M2090
G3D Mark Score
1,452+4%
1,400
Architecture
Kepler
Fermi 2.0
Process Node
28 nm
40 nm
Shading Units
384
512+33%
Compute (TFLOPS)
0.7296 TFLOPS
1.332 TFLOPS+83%
ROPs
16
48+200%
TMUs
32
64+100%
L1 Cache
0.03 MB
1 MB+3233%
L2 Cache
256 KB
768 KB+200%

Advanced Features (DLSS/FSR)

The GeForce GTX 660M gets NVIDIA DLSS, which still tends to look cleaner in motion. The Tesla M2090 leans on FSR, which is flexible and widely supported, but usually a bit rougher at the same settings.

FeatureGeForce GTX 660MTesla M2090
Upscaling Tech
Upscaling support
Upscaling support
Frame Generation
Not Supported
Not Supported
Ray Reconstruction
No
No
Low Latency
NVIDIA Reflex
Standard
💾

Video Memory (VRAM)

The GeForce GTX 660M has 2 GB of VRAM, while the Tesla M2090 carries 512 MB. GeForce GTX 660M gives you 300% more memory capacity, which matters more once you move into heavier textures, mods, or higher resolutions. Memory bus width is 128-bit on the GeForce GTX 660M and 64-bit on the Tesla M2090. L2 Cache: 256 KB (GeForce GTX 660M) vs 768 KB (Tesla M2090) — the Tesla M2090 has significantly larger on-die cache to reduce VRAM reliance.

FeatureGeForce GTX 660MTesla M2090
VRAM Capacity
2 GB+300%
0.5 GB
Memory Type
GDDR5
GDDR5
Bus Width
128-bit+100%
64-bit
L2 Cache
256 KB
768 KB+200%
🖥️

Display & API Support

DirectX support: 12 (11_0) (GeForce GTX 660M) vs 12 (FL 11_0) (Tesla M2090). OpenGL: 4.5 vs 4.6. Maximum simultaneous displays: 4 vs 0.

FeatureGeForce GTX 660MTesla M2090
DirectX
12 (11_0)
12 (FL 11_0)
OpenGL
4.5
4.6+2%
Max Displays
4
0
🎬

Media & Encoding

Supported codecs: H.264,VC-1,MPEG-2,MPEG-4 ASP (GeForce GTX 660M) vs CUDA,OpenCL (Tesla M2090).

FeatureGeForce GTX 660MTesla M2090
Encoder
NVENC 1st Gen
Decoder
PureVideo HD VP5
Codecs
H.264,VC-1,MPEG-2,MPEG-4 ASP
CUDA,OpenCL
🔌

Power & Dimensions

The GeForce GTX 660M draws 50W versus the Tesla M2090's 250W — a 133.3% difference. The GeForce GTX 660M is more power-efficient. Recommended PSU: 350W (GeForce GTX 660M) vs 350W (Tesla M2090). Power connectors: 1x 6-pin vs PCIe-powered.

FeatureGeForce GTX 660MTesla M2090
TDP
50W-80%
250W
Recommended PSU
350W
350W
Power Connector
1x 6-pin
PCIe-powered
Length
248mm
Height
111mm
Slots
0-100%
2
Temp (Load)
60°C
Perf/Watt
29.0+418%
5.6
💰

Value Analysis

The newer card here is GeForce GTX 660M (2012 vs 2011).

FeatureGeForce GTX 660MTesla M2090
MSRP
$2500
Codename
GK107
GF110
Release
March 22 2012
July 25 2011
Ranking
#774
#530

Affiliate Disclosure

ChipVERSUS is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com. We may earn a commission on qualifying purchases made through our links. This comes at no additional cost to you and helps support our work in providing comprehensive PC building guides and tools.

Amazon and the Amazon logo are trademarks of Amazon.com, Inc. or its affiliates.